Qwen3.5 397B A17B is cheaper than GPT-5.4 at $0.6/M vs $2.5/M input tokens.
| Spec | GPT-5.4 | Qwen3.5 397B A17B |
|---|---|---|
| Input price | $2.5/M tokens | $0.6/M tokens |
| Output price | $15/M tokens | $3.6/M tokens |
| Context window | 1.1M tokens | 262K tokens |
| Released | Mar 2026 | Feb 2026 |
Not enough votes to call it. On the specs, GPT-5.4 has the edge: bigger context window, major provider backing.
Qwen3.5 397B A17B costs 4.2x less per token.
Qwen3.5 397B A17B is cheaper on both — 4.2× input, 4.2× output
Qwen3.5 397B A17B uses 22.8x more emoji
Qwen3.5 397B A17B is cheaper than GPT-5.4 at $0.6/M vs $2.5/M input tokens.
| Spec | GPT-5.4 | Qwen3.5 397B A17B |
|---|---|---|
| Input price | $2.5/M tokens | $0.6/M tokens |
| Output price | $15/M tokens | $3.6/M tokens |
| Context window | 1.1M tokens | 262K tokens |
| Released | Mar 2026 | Feb 2026 |
Not enough votes to call it. On the specs, GPT-5.4 has the edge: bigger context window, major provider backing.
Qwen3.5 397B A17B costs 4.2x less per token.
Qwen3.5 397B A17B is cheaper on both — 4.2× input, 4.2× output
Qwen3.5 397B A17B uses 22.8x more emoji